Combined treatment with testosterone (T) + estradiol-17beta (E2) induces prostate adenocarcinomas in 100 percent of NBL rats, and T alone a 40 percent incidence. This is the only model in which steroid hormones alone produce a high prostate cancer incidence; it offers opportunities to study the mechanisms whereby steroid hormones cause prostate cancer. We hypothesize that E2, either formed endogenously from T by the enzyme aromatase or exogenously administered, acts via a genotoxic mechanism that involves the generation of reactive oxygen species to produce prostate carcinomas in the presence of T, and that estrogen receptor (ER)-mediated mechanisms are not involved. To address the hypothesis and resolve some uncertainties about the model, the following specific aims are proposed. Aim 1: To establish whether E2 is an essential factor in the hormonal induction of prostate cancer by T, and whether it acts via a mechanism that is ER-independent; we will establish whether treatment with an aromatase inhibitor reduces or eliminates prostate cancer induction by T, and to demonstrate that antiestrogen treatment does not affect the tumor response. Aim 2: To determine whether induction of oxidative DNA damage and of prostate carcinomas by T and E2 are linked and, therefore, probably causally related; we will administer antioxidant vitamins (C and E) to T+E2-treated rats, which should reduce both prostate cancer induction and prostatic oxidative DNA damage (measured by formation of 8-hydroxydeoxyguanosine). Aim 3: To demonstrate that T+E2-induced carcinomas have the potential to metastasize and determine whether T+E2-induced peripheral prostate dysplasia can progress to cancer; we will determine whether lowering of the E2 dose or shortening of the E2 treatment duration, while continuing T administration, will sufficiently lower or delay pituitary tumor formation and prolong survival to allow progression of prostate carcinomas to the metastatic stage. Aim 4: To establish whether T+E2- induced prostate carcinomas are androgen-sensitive and their induction is androgen-dependent and prolactin-independent; we will test the response of prostate cancers transplanted into syngeneic rats to castration and determining the effect of castration of T+E2-treated rats at a time at which very small carcinomas have already formed.